ANC Workshop - Alessandro Fontanella, Bryan Li

Tuesday, 28th March 2023

Methods for classification and anomaly detection in medical imaging - Alessandro Fontanella

Abstract: In some medical imaging tasks and other settings where only small parts of the image are informative for the classification task, traditional CNNs can sometimes struggle to generalise. Manually annotated Regions of Interest (ROI) are often used to isolate the most informative parts of the image. However, these are expensive to collect and may vary significantly across annotators. To overcome these issues, we propose a framework that employs saliency maps to obtain soft spatial attention masks that modulate the image features at different scales. We show that our approach increases the baseline classification accuracy of lesions in brain and lung CT scans. We investigate the best way to generate the saliency maps employed in our architecture and propose a way to obtain them from adversarially generated counterfactual images. Finally, we introduce a way to employ these saliency maps, together with diffusion models, to generate a healthy version of a diseased image and use it to obtain an anomaly map of the pathological areas.

V1T: large-scale mouse V1 response prediction using a Vision Transformer - Bryan Li

Abstract: Accurate predictive models of the visual cortex neural response to natural visual stimuli remain a challenge in computational neuroscience. In this work, we introduce V1T, a novel Vision Transformer based architecture that learns a shared visual and behavioral representation across animals. We evaluate our model on two large datasets recorded from mouse primary visual cortex and outperform previous convolution-based models by more than 12.7% in prediction performance. Moreover, we show that the attention weights learned by the Transformer correlate with the population receptive fields. Our model thus sets a new benchmark for neural response prediction and captures characteristic features of the visual cortex.

Event type: Workshop

Date: Tuesday, 28th March 2023

Time: 11:00

Location: G.03

Speaker(s): Alessandro Fontanella, Bryan Li

Chair/Host: Patricia Rubisch